In semi-conductor manufacturing, chips, after fabrication, are usually probed to check their electrical characteristics. As the density of the input/output pads on the chips increases such probing of the chips becomes more and more difficult.
One type of prior art probe is formed on the surface of a substrate typically formed of a semiconductor, such as silicon, or an insulating material. The probe is comprised of a plurality of metallic lines connecting contacts or electrodes, in the form of bumps or tips, to respective signal pads connected to an external tester or other signal source. The probe tips are positioned to contact the Input/Output (I/O) pads on a chip to be tested.
Probes, so formed, were found to be unacceptable for they are rigid and non-compliant and fail to compensate for differences in height between the I/O pads on the chip or wafer being tested and the rigid probe tips.
In an attempt to overcome this difficulty such probes can be over driven, that is, forced into the chip pads, in order to assure good contact between the chip pads and probe contacts. This over driving of the probe, however, was found to significantly damage both the probe tips and the chip it was contacting thus increasing the possibility of damage to both the probe and the chip being tested resulting in poor chip test yields and damaged probes.
In an attempt to overcome the flexibility problem, B. Leslie, et al., described, in an article entitled "Membrane Probe Card Technology for VLSI Wafer Testing" which was given at the 1988 International Test Conference, pp. 601-607, a new probe in which both the contact electrodes and interconnecting wiring were formed on a flexible film. It was believed that by applying pressure from a side opposite to the contact electrodes improved contact between a pad and a probe would be achieved. A similar device was disclosed in an article by M. Beiley, et al.; entitled "Array Probe Card," Multi-Chip Module Conference, pp. 28-31, 1992 and in U.S. Pat. No. 5,103,557 by G. J. Leedy, et al.
However, the probes, described in these articles, were found to also have contact electrode pressure problems. It was found that uniform contact electrode pressure on the pads could not be made due to the excessive flexibility of the entire film. In addition, probes formed from flexible film as described in the papers and patent cannot achieve the wiring density necessary to test high I/O chips.
Still another probe arrangement is described in patent application Ser. No. 08/321,503, now U.S. Pat. No. 5,625,298, which is assigned to the same assignee as the present invention. In this invention a small cavity was formed in the underlying substrate and a thin elastic film formed thereover. However, this method suffers from severe problems in forming the film across the mouth of the cavity. For this reason it was found to be unsuitable for use in a manufacturing or mass production setting.
Thus, there still exists a need for a probe which overcomes the manufacturing and contact problems encountered by the prior art and which can be readily adaptable for both individual chip or full wafer probing and can be easily, quickly and inexpensively produced for use in manufacturing as well as being provided with test circuitry.